China

Large Language Models (LLMs) have demonstrated impressive capabilities in reasoning, positioning them as promising tools for supporting human problem-solving. However, what happens when their performance is affected by *misinformation*, i.e., incorrect inputs introduced by users due to oversights or gaps in knowledge? Such misinformation is prevalent in real-world interactions with LLMs, yet how it propagates within LLMs’ reasoning process remains underexplored. Focusing on mathematical reasoning, we present a comprehensive analysis of how misinformation affects intermediate reasoning steps and final answers. We also examine how effectively LLMs can correct misinformation when explicitly instructed to do so. Even with explicit instructions, LLMs succeed less than half the time in rectifying misinformation, despite possessing correct internal knowledge, leading to significant accuracy drops (10.02% – 72.20%). Further analysis shows that applying factual corrections early in the reasoning process most effectively reduces misinformation propagation, and fine-tuning on synthesized data with early-stage corrections significantly improves reasoning factuality. Our work offers a practical approach to mitigating misinformation propagation.

EMNLP 2025

Unraveling Misinformation Propagation in LLM Reasoning

poster

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community."<br>

*Message from the General Chair, Dirk Hovy*

[**Link to Conference Handbook**](https://drive.google.com/file/d/1johU5QqVVYO4RfH7QcIORr7qrVBdzdwC/view?usp=sharing)





<br>

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

Large language models (LLMs) excel in general tasks but struggle with domain-specific ones, requiring fine-tuning with specific data. With many open-source LLMs available, selecting the best model for fine-tuning downstream tasks is challenging, primarily focusing on how to quickly identify the optimal LLM. We introduce a Data and Model Compression Framework (DaMoC) that addresses this challenge by: 1) Data Level: A systematic categorization of data filtering methodologies for LLMs is first established, classifying them into three distinct paradigms: (1) distribution-aware methods, (2) quality-aware methods, and (3) hybrid approaches considering both dimensions. Further, we enhance the density of key tokens in the text achieving token compression. Subsequently, we use an LLM to iterative rewrite the text to optimize its expression. 2) Model Level: We use layer similarity scores to assess each layer's importance and remove those with lower importance. Then, we introduce a sparse merging paradigm to preserve as much of the original model's capability as possible. Extensive experiments on four datasets, medical Q&A, financial Q&A, general Q&A, and reading comprehension, show that we can select the optimal LLM while saving approximately 20-fold in training time.

DaMoC: Efficiently Selecting the Optimal Large Language Model for Fine-tuning Domain Tasks Based on Data and Model Compression

Mixture-of-Experts (MoE) models have become increasingly powerful in multimodal learning by enabling modular specialization across modalities. However, their effectiveness remains unclear when additional modalities introduce more noise than complementary information. Existing approaches, such as the Partial Information Decomposition, struggle to scale beyond two modalities and lack the resolution needed for instance-level control. We propose **B**eyond **T**wo-modality **W**eighting (**BTW**), a bi-level, non-parametric weighting framework that combines instance-level Kullback-Leibler (KL) divergence and modality-level mutual information (MI) to dynamically adjust modality importance during training. Our method does not require additional parameters and can be applied to an arbitrary number of modalities. Specifically, BTW computes per‐example KL weights by measuring the divergence between each unimodal and the current multimodal prediction, and modality‐wide MI weights by estimating global alignment between unimodal and multimodal outputs. Extensive experiments on sentiment regression and clinical classification demonstrate that our method significantly improves regression performance and multiclass classification accuracy.

BTW: A Non-Parametric Variance Stabilization Framework for Multimodal Model Integration

Conversational derailment — when online discussions stray from their intended topics due to toxic or inappropriate remarks — is a common issue on online platforms. These derailments can have negative impacts on users and the online community. While previous work has focused on post hoc identification of toxic content, recent efforts emphasize proactive prediction of derailments before they occur, enabling early moderation. However, forecasting derailment is difficult due to the context-dependent emergence of toxicity and the need for timely alerts. We prompt pre-trained large language models (LLMs) to predict conversational derailment without task-specific fine-tuning. We compare a range of prompting strategies, including chain-of-thought reasoning (CoT) and few-shot exemplars, across small and large scale models, and evaluate their performance and inference-cost trade-offs on derailment benchmarks. Our experiments show that the best prompting configuration attains state-of-the-art performance, and forecasts derailments earlier than existing approaches. These results demonstrate that LLMs, even without fine-tuning, can serve as an effective tool for proactive conversational moderation.

Can LLMs Be Efficient Predictors of Conversational Derailment?

Longitudinal experiential data offers rich insights into dynamic human states, yet building models that generalize across diverse contexts remains challenging. This paper addresses how to best represent multi-modal longitudinal experiential data as text and formulate prediction tasks to maximize large language model (LLM) cross-distribution generalization. We propose ConText-LE, a framework grounded in linguistic and cognitive theories of contextual meaning-making, which systematically investigates text representation strategies and output formulations for robust behavioral forecasting. Our novel Meta-Narrative representation synthesizes complex temporal patterns into semantically rich narratives, while Prospective Narrative Generation reframes prediction as a generative task aligned with LLMs' inherent contextual understanding capabilities. Through comprehensive experiments on three diverse longitudinal datasets, we address the critical but underexplored challenge of cross-distribution generalization in mental health and educational behavior forecasting. We demonstrate that combining Meta-Narrative input with Prospective Narrative Generation significantly outperforms existing LLM-based approaches, achieving up to 12.28% improvement in out-of-distribution accuracy and up to 11.99% improvement in F1 scores over binary classification methods. Bidirectional evaluation and architectural ablation studies confirm the robustness of our approach, establishing ConText-LE as an effective framework for developing reliable behavioral forecasting systems across temporal and contextual shifts.

ConText-LE: Cross-Distribution Generalization for Longitudinal Experiential Data via Narrative-Based LLM Representations

Recent years have witnessed a growing interest in personalizing the responses of large language models (LLMs). While existing evaluations primarily focus on whether a response aligns with a user’s preferences, we argue that factuality is an equally important yet often overlooked dimension. In the context of personalization, we define a model as robust if its responses are both factually accurate and align with the user preferences. To assess this, we introduce PERG, a scalable framework for evaluating robustness in LLMs, along with a new dataset, PERGData. We evaluate twelve state-of-the-art models across four prompting strategies. Our findings show that current LLMs struggle with robust personalization: even the strongest model (LLaMA3-70B) fails to maintain correctness in 5% of previously successful cases without personalization, while smaller models (e.g., 7B-scale) can fail more than 20% of the time. Further analysis reveals that robustness is significantly affected by the nature of the query and the type of user preference. To mitigate these failures, we propose Pref-Aligner, a two-stage approach that improves robustness by an average of 25% across models. Our work highlights critical gaps in current evaluation practices and introduces tools and metrics to support more reliable, user-aligned LLM deployments.

Benchmarking and Improving LLM Robustness for Personalized Generation

Large Language Models (LLMs) have shown significant limitations in understanding creative content, as demonstrated by Hessel et al. (2023)'s influential work on the New Yorker Cartoon Caption Contest (NYCCC). Their study exposed a substantial gap between LLMs and humans in humor comprehension, establishing that understanding and evaluating creative content is key challenge in AI development. We revisit this challenge by decomposing humor understanding into three components and systematically improve each: enhancing visual understanding through improved annotation, utilizing LLM-generated humor reasoning and explanations, and implementing targeted alignment with human preference data. Our refined approach achieves 82.4% accuracy in caption ranking, significantly improving upon the previous 67% benchmark and matching the performance of world-renowned human experts in this domain. Notably, while attempts to mimic subgroup preferences through various persona prompts showed minimal impact, model finetuning with crowd preferences proved remarkably effective. These findings reveal that LLM limitations in creative judgment can be effectively addressed through focused alignment to specific subgroups and individuals. Lastly, we propose the position that achieving artificial general intelligence necessitates systematic collection of human preference data across creative domains. We advocate that just as human creativity is deeply influenced by individual and cultural preferences, training LLMs with diverse human preference data may be essential for developing true creative understanding.

Bridging the Creativity Understanding Gap: Small-Scale Human Alignment Enables Expert-Level Humor Ranking in LLMs

Alignment of large language models with explicit principles (such as helpfulness, honesty, and harmlessness) is crucial for ensuring safe and reliable AI systems. However, standard reward-based alignment methods typically collapse diverse feedback into a single scalar reward, entangling multiple objectives into one opaque training signal, which hinders interpretability. In this work, we introduce QA-LIGN, an automatic symbolic reward decomposition approach that preserves the structure of each constitutional principle within the reward mechanism. Instead of training a black-box reward model that outputs a monolithic score, QA-LIGN formulates principle-specific evaluation questions and derives separate reward components for each principle, making it a drop-in reward model replacement. Experiments aligning an uncensored large language model with a set of constitutional principles demonstrate that QA-LIGN offers greater transparency and adaptability in the alignment process. At the same time, our approach achieves performance on par with or better than a DPO baseline. Overall, these results represent a step toward more interpretable and controllable alignment of language models, achieved without sacrificing end-task performance.

QA‑LIGN: Aligning LLMs through Constitutionally Decomposed QA

Controlled paraphrase generation produces rephrasings that preserve meaning while enabling targeted stylistic and linguistic modifications. We introduce LingConv, an encoder-decoder framework supporting fine-grained control over 40 linguistic attributes in English. A novel inference-time quality control mechanism iteratively refines attribute embeddings, yielding paraphrases that closely match target properties without sacrificing semantic fidelity. LingConv reduces attribute error by up to 34% over baselines, with quality control providing an additional 14% improvement.

Linguistically-Controlled Paraphrase Generation

Large language models (LLMs) are ubiquitous in modern day natural language processing. However, previous work has shown degraded LLM performance for under-represented English dialects. We analyze the effects of typifying "standard" American English language questions as non-"standard" dialectal variants on multiple choice question answering tasks and find up to a 20% reduction in accuracy. Additionally, we investigate the grammatical basis of under-performance in non-"standard" English questions. We find that individual grammatical rules have varied effects on performance, but some are more consequential than others: three specific grammar rules (existential “it,” zero copula, and y'all) can explain the majority of performance degradation observed in multiple dialects. We call for future work to investigate bias mitigation methods focused on individual, high-impact grammatical structures.

Analyzing Dialectical Biases in LLMs for Knowledge and Reasoning Benchmarks

We introduce a Neural-Symbolic Task Planning framework integrating Large Language Model (LLM) decomposition with category-theoretic verification for resource-aware, temporally consistent planning. Our approach represents states as objects and valid operations as morphisms in a categorical framework, ensuring constraint satisfaction through mathematical pullbacks. We employ bidirectional search that simultaneously expands from initial and goal states, guided by a learned planning distance function that efficiently prunes infeasible paths. Empirical evaluations across three planning domains demonstrate that our method outperforms existing baselines by up to 26.1% in completion rates while reducing relative resource violation rate by up to 77%. These results highlight the synergy between LLM-based operator generation and category-theoretic verification for reliable planning in domains requiring both resource-awareness and temporal consistency.

Downloads

Next from EMNLP 2025

DaMoC: Efficiently Selecting the Optimal Large Language Model for Fine-tuning Domain Tasks Based on Data and Model Compression